Information Processing Apparatus, Recording Medium, and Information Processing Method

Abstract

Provided are an information processing apparatus, a recording medium, and an information processing method. The information processing apparatus includes: a first extraction unit that extracts, from the result of character recognition of a read image of a document including items and values corresponding to the items, a value corresponding to a first item that is an item representing a deadline; a generation unit that, when the value extracted by the first extraction unit is a first string other than a date format, generates a second string in a date format representing the date of the deadline according to the result of the character recognition; and an output unit that outputs the generated second string as a value corresponding to the first item.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, a recording medium, and an information processing method. Background Art

[0002] In Patent Document 1, there is disclosed a document production apparatus that can produce document information including information such as a date. The document production apparatus includes: a date information storage unit that stores date information and information on the representation form of the date; an extraction unit that extracts information consistent with the representation form information stored in the date information storage unit from the produced document information; a unit that cuts out date-related information from the information extracted by the extraction unit and compares it with the date information in the date information storage unit; and a unit that displays and outputs the inconsistent part when the cut-out information is inconsistent with the date information by this unit.

[0003] In Patent Document 2, there is disclosed an information retrieval apparatus that includes: an input unit that inputs various data; a registration unit that registers each data by assigning retrieval information; a storage unit that stores the registered information group; a display unit that displays and outputs various data; an input date retrieval keyword setting unit that sets the date of the input data as the input date retrieval keyword; a retrieval unit that retrieves the information group according to the input date retrieval keyword; and a conversion unit that, when a string related to a date is input, converts this into date information and retrieves the input date of the data according to the string related to the date.

[0004] In Patent Document 3, there is disclosed a system for managing documents that includes: a cut processing unit that cuts out a string that becomes an index for the input document together with a retrieval category according to cut rule data; a plurality of shaping processing units that each have a rule for converting a string into a formal representation and convert the string according to each rule; and a shaping processing control unit that, for the string cut out by the cut processing unit together with the retrieval category, calls the corresponding shaping processing unit according to the corresponding shaping rule data, thereby gives an instruction to replace it with a formal representation, and outputs the result as a shaped document.

[0005] Patent Document 1: Japanese Patent Laid-Open No. 08-069451

[0006] Patent Document 2: Japanese Patent Laid-Open No. 2002-197105

[0007] Patent Document 3: Japanese Patent Laid-Open No. 2004-185452

[0008] There are documents such as forms that contain items and values corresponding to the items. And sometimes, for items that require a date as the value, such as the delivery deadline, there are values other than dates. When a person reads the document, they can usually infer the date from the values other than the date. However, in the character recognition of the read image of the document, date conversion is not performed. Summary of the Invention

[0009] An object of the present invention is to provide an information processing device, a recording medium, and an information processing method. In the information processing device, when a string other than the date format is extracted as the value for an item representing a deadline from the result of character recognition of a read image of a document, a string in the date format representing the date of the deadline can be output.

[0010] The information processing device according to the first aspect includes: a first extraction unit that extracts, from the result of character recognition of a read image of a document including items and values corresponding to the items, a value corresponding to a first item that is an item representing a deadline; a generation unit that, when the value extracted by the first extraction unit is a first string other than the date format, generates, based on the character recognition result, a second string in the date format representing the date of the deadline; and an output unit that outputs the generated second string as the value corresponding to the first item.

[0011] In the information processing device according to the second aspect, in the information processing device according to the first aspect, the generation unit includes: an acquisition unit that acquires a second item different from the first item; and a second extraction unit that extracts, from the character recognition result, a value corresponding to the second item, and generates the second string in the date format based on the value extracted corresponding to the acquired second item.

[0012] In the information processing device according to the third aspect, in the information processing device according to the second aspect, the second item is an item representing the starting date that is the reference for calculating the deadline. When the value extracted by the second extraction unit is a third string in the date format, the generation unit acquires the date represented by the third string as the date of the starting date.

[0013] In the information processing device according to the fourth aspect, in the information processing device according to the third aspect, when the value extracted by the second extraction unit is a fourth string other than the date format, the generation unit acquires the date represented by the fifth string described in a predetermined date description area of the image as the date of the starting date.

[0014] In the information processing apparatus according to the fifth mode, in the information processing apparatus according to the third mode or the fourth mode, the acquisition unit acquires, from the item storage unit, an item indicating the starting date corresponding to the type of the specified document, and the item storage unit stores in advance, for each type of a plurality of documents, an item indicating the starting date.

[0015] In the information processing apparatus according to the sixth mode, in the information processing apparatus according to the third mode or the fourth mode, the acquisition unit acquires an item indicating the starting date by performing morpheme analysis on the first string.

[0016] In the information processing apparatus according to the seventh mode, in any one of the information processing apparatuses according to the third mode to the sixth mode, the generation unit uses a calculation formula for calculating the date of the deadline from the date of the starting date, acquires the date of the deadline corresponding to the acquired date of the starting date, and generates the second string in the date format.

[0017] In the information processing apparatus according to the eighth mode, in the information processing apparatus according to the seventh mode, the calculation formula for calculating the date of the deadline is acquired from a calculation formula storage unit according to the representation pattern of the first string, and the calculation formula storage unit stores in advance calculation formulas for a plurality of representation patterns other than the date format.

[0018] In the information processing apparatus according to the ninth mode, in any one of the information processing apparatuses according to the third mode to the sixth mode, the generation unit acquires the period from the starting date to the deadline by performing morpheme analysis on the first string, converts the acquired period into days, acquires the date of the deadline corresponding to the number of days and the date of the starting date, and thereby generates the second string in the date format.

[0019] In the information processing apparatus according to the tenth mode, in any one of the information processing apparatuses according to the first mode to the ninth mode, the output unit displays the second string in the date format on a display unit.

[0020] In the information processing apparatus according to the eleventh mode, in the information processing apparatus according to the tenth mode, the output unit replaces the first string other than the date format included in the result of the character recognition with the second string in the date format.

[0021] A recording medium according to the twelfth mode records a program that causes a computer to function as each unit of the information processing apparatus according to any one of the first mode to the tenth mode.

[0022] The information processing method according to the 13th mode includes the following steps: a first extraction step of extracting, from the result of character recognition of a read image of a document including items and values corresponding to the items, a value corresponding to a first item that is an item representing a deadline; a generation step of generating, when the value extracted by the first extraction step is a first string other than a date format, a second string in a date format representing the date of the deadline according to the character recognition result; and an output step of outputting the generated second string as a value corresponding to the first item.

[0023] Advantages of the Invention

[0024] According to the 1st mode, the 12th mode, and the 13th mode, when a string other than a date format is extracted from the result of character recognition of a read image of a document as a value for an item representing a deadline, a string in a date format representing the date of the deadline can be output.

[0025] According to the 2nd mode, a string in a date format representing the date of the deadline can be generated based on a value corresponding to an item different from the item representing the deadline in the result of character recognition.

[0026] According to the 3rd mode, the date of the starting date can be obtained from the value corresponding to the item representing the starting date in the result of character recognition.

[0027] According to the 4th mode, the date of the starting date can be obtained from the string described in a date description area determined in advance in the result of character recognition.

[0028] According to the 5th mode, the item representing the starting date can be obtained according to the type of the document.

[0029] According to the 6th mode, the item representing the starting date can be obtained from the string other than the extracted date format.

[0030] According to the 7th mode and the 8th mode, the date of the deadline can be obtained from the date of the starting date and a calculation formula without calculating the number of days.

[0031] According to the 9th mode, the date of the starting date and the number of days can be obtained from the string other than the extracted date format.

[0032] According to the 10th mode, the generated string in a date format can be displayed to the user.

[0033] According to the 11th mode, the result of character recognition can be corrected so that the generated string in a date format corresponds to the item representing the deadline. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Embodiments of the present invention will be described in detail with reference to the following drawings.

[0035] Figure 1 is a schematic diagram showing an example of a document that is an object of character recognition;

[0036] Figure 2 is a schematic diagram showing an example of a screen displayed to a user;

[0037] Figure 3 is a schematic diagram showing another example of a screen displayed to a user;

[0038] Figure 4 is a diagram illustrating a representation mode of the date format of a deadline;

[0039] Figure 5 is a diagram illustrating a representation mode other than the date format of a deadline;

[0040] Figure 6 is a block diagram showing the electrical structure of an information processing apparatus according to the first embodiment;

[0041] Figure 7 is a chart showing a calculation formula table;

[0042] Figure 8 is a chart showing a starting date item list;

[0043] Figure 9 is a functional block diagram showing the functional structure of an information processing unit according to the first embodiment;

[0044] Figure 10 is a flowchart showing the process flow based on the character recognition program according to the first embodiment;

[0045] Figure 11 is a block diagram showing the structure of a storage unit of an information processing apparatus according to the second embodiment;

[0046] Figure 12 is a flowchart showing the process flow based on the character recognition program according to the second embodiment;

[0047] Figure 13 is a flowchart showing the process flow of the analysis process of string A;

[0048] Figure 14 is a flowchart showing the process flow of the date extraction process of the starting date;

[0049] Figure 15 is a flowchart showing the process flow of the analysis process of string B;

[0050] Figure 16 is a diagram explaining the date conversion rule.

[0051] Symbol Explanation

[0052] 10 - Information processing device, 12 - Information processing unit, 14 - Operation display unit, 16 - Image reading unit, 18 - Communication unit, 20 - Storage unit, 22 - Character recognition program, 24 - Expiration date display mode list, 26 - Calculation formula table, 28 - Starting date item list, 30 - Character recognition unit, 32 - Character recognition result storage unit, 34 - Extraction unit, 36 - Form determination unit, 38 - Date generation unit, 40 - Output unit, 50 - Morpheme analysis dictionary, 52 - Days conversion rule. Detailed Embodiment

[0053] Hereinafter, with reference to the accompanying drawings, an example of an embodiment of the present invention will be described in detail.

[0054] - First Embodiment -

[0055] (Document to be the object of character recognition)

[0056] First, the document that is the object of character recognition in this embodiment will be described. Figure 1 It is a schematic diagram showing an example of a document that is the object of character recognition. The document that is the object of character recognition is a document containing items and values corresponding to the items. For example, in forms such as estimate sheets, values corresponding to each item are described. Hereinafter, the case where the document is an estimate sheet will be described.

[0057] As Figure 1 shown, in the estimate sheet, items include management number, issue date, estimated amount, payment deadline, estimated validity period, product name, unit price, quantity, amount, etc. For example, for the item "issue date", the value "January 7, 2019" is described. Moreover, for the item "estimated validity period", the value "1 month from the estimate date" is described.

[0058] Taking the estimate sheet as the original document, the image of the estimate sheet is read. In the page of the read image, an image representing the value corresponding to the item is arranged near the image representing the item. If character recognition is performed on the read image, character recognition is performed for each image in the page. The character recognition result includes the position of the image representing the item in the page, the string representing the item, the position of the image representing the value in the page, and the string representing the value. The position in the page can be represented by position coordinates with a predetermined point as the origin.

[0059] (Items and Values)

[0060] The string representing an item and the string representing the value corresponding to the item are associated according to their respective positions within the page. By specifying the string representing the item, the string representing the corresponding value is extracted from the result of character recognition. Hereinafter, the string representing the item is referred to as the "keyword string", and the string representing the value is referred to as the "value string". Moreover, the process of specifying the keyword string to extract the value string is referred to as "keyword-value extraction".

[0061] Figure 2 It is a schematic diagram showing an example of the screen displayed to the user. Figure 2 The screen shown is a setting screen for setting the conditions of character recognition. The user operates the setting screen to select an item and the type of document. For example, in the illustrated example, as the type of document, "estimate" is specified. Moreover, the item of the estimate expiration date is selected, and the string "estimate expiration date" is specified as the keyword string. If the start button is pressed to indicate the start of the character recognition process, the character recognition process for reading the image starts. Then, keyword-value extraction is performed on the result of the character recognition to extract the value string.

[0062] (String in date format)

[0063] An item representing a "period" such as the issue date or the estimate expiration date is an item that requires a string in date format representing the date of the period as the corresponding value. When performing keyword-value extraction with "estimate expiration date" as the keyword string, it is assumed that a value string in date format will be extracted.

[0064] However, for an item representing a period, as the value, there may sometimes be associated a string other than in date format. For example, Figure 1 In the illustrated example, for the keyword string of "estimate expiration date", the string "1 month from the estimate date" is extracted as the value string. A string other than in date format does not directly represent the date of the period.

[0065] Figure 4 It is a diagram illustrating the representation patterns of the date format of the period. As Figure 4 shown, as the representation patterns of the date format, 2018 / 10 / 30, October 30, 2018 (Heisei 30), 2018-10-30, 2018.10.30, Oct 30, 2018, October 30, 2018 (H30), etc. can be cited.

[0066] When representing a date in the Gregorian calendar, the string in date format is a string in which a 4-digit number representing the "year", the text of the "year", a 1- or 2-digit number representing the "month", the text of the "month", a 1- or 2-digit number representing the "day", and the text of the "day" are arranged in the order of year → month → day. It is also possible to replace the text of the year, month, and day with marks such as slashes, hyphens, and dots. Depending on the language, the order of recording the year, month, and day can be swapped, or the text of the year, month, and day can be replaced with letters.

[0067] When representing a date in the Japanese calendar, the string in date format is a string in which the text of the era name, a 1- or 2-digit number representing the "year", the text of the "year", a 1- or 2-digit number representing the "month", the text of the "month", a 1- or 2-digit number representing the "day", and the text of the "day" are arranged in the order of year → month → day. It is also possible to replace the text of the era name with the first letter of the alphabet.

[0068] In the present embodiment, a string in a pattern of date format that is predetermined in advance, such as a string including numbers representing the "year", numbers representing the "month", and numbers representing the "day", is set as the "string in date format".

[0069] A string other than the date format is a string other than the above-mentioned "string in date format". Figure 5 It is a diagram showing a representation pattern other than the date format of the exemplified term. As Figure 5 shown, as representation patterns other than the date format of the term, examples include 3 months, 30 days, 1 month, 1 month from now, 1 month from the valuation date, 1 month from the issue date, within 1 month from the issue date, 1 month from the valuation date, 30 days from the issue date of this form, 3 months from the valuation preparation date, 30 days from the issue date of this valuation form, from November 8, 2013 to November 30, until November 30, 2018, 30 days after issue, 3 months after issue, 2 weeks after valuation, 1 month after submitting this valuation, the last day of July 2013, etc.

[0070] As Figure 5 shown, a string other than the date format is, for example, a string that does not include numbers representing the "year", numbers representing the "month", and numbers representing the "day", or a string that includes extra text after the string in date format. Extra text is, for example, a particle or a case particle (~from) or a conjunction (~to~), etc.

[0071] (Generation of the string in date format)

[0072] In the present embodiment, when a string other than a date format is associated as a value with an item representing a time limit, a string in a date format representing the date of the time limit is generated using the result of character recognition. The date of the time limit is calculated based on the date of the start date. For example, when the date of the start date can be obtained from the result of character recognition, the date of the time limit is obtained from the date of the start date, and a string in a date format representing the date of the time limit is generated. Then, the generated string in the date format is displayed to the user.

[0073] Figure 3 It is a schematic diagram showing an example of a screen displayed to the user. Figure 3 The screen shown represents the result of extracting the keyword value with "appraisal validity period" as the keyword string. In the illustrated example, a string other than the date format of "1 month from the appraisal date" is extracted as the value string.

[0074] In the present embodiment, the generated string in the date format is displayed to the user together with or instead of the extracted value string. For example, both the extracted value string and the generated string in the date format may be displayed as options for the user to select. Moreover, the string in the date format other than the date format may be replaced with the generated string in the date format to correct the result of character recognition.

[0075] <Information Processing Device>

[0076] Next, the hardware structure of the information processing device will be described.

[0077] Figure 6 It is a block diagram showing the electrical structure of the information processing device according to the first embodiment. As Figure 6 shown, the information processing device 10 includes an information processing unit 12 which is a computer that controls each connected part and performs various operations. That is, the information processing unit 12 includes a CPU (Central Processing Unit) 12A, a ROM (Read Only Memory) 12B, a RAM (Random Access Memory) 12C, a non-volatile memory 12D, and an input / output unit (I / O) 12E.

[0078] The CPU 12A, the ROM 12B, the RAM 12C, the memory 12D, and the I / O 12E are respectively connected via a bus 12F. The CPU 12A reads a program stored in a storage device such as the ROM 12B and uses the RAM 12C as a working area to execute the program.

[0079] The information processing apparatus 10 includes, for example, an operation display unit 14, an image reading unit 16, a communication unit 18, and a storage unit 20. The operation display unit 14, the image reading unit 16, the communication unit 18, and the storage unit 20 are respectively connected to the I / O 12E of the information processing unit 12.

[0080] The operation display unit 14 displays various screens to the user and accepts operations from the user. The operation display unit 14 is configured to include, for example, a touch panel. The image reading unit 16 is a device that reads an image of a set original document. The communication unit 18 is an interface for communicating with an external device via a wired or wireless communication line. The storage unit 20 is an external storage device such as a hard disk.

[0081] Various programs and various data are stored in a storage device such as the ROM 12B. The storage area for the programs is not limited to the ROM 12B. Various programs can be stored in other storage devices such as the memory 12D or the storage unit 20, or can be obtained from an external device via the communication unit 18.

[0082] Moreover, various drives can be connected to the information processing unit 12. The various drives are devices that read data from or write data to a computer-readable and portable recording medium such as a CD-ROM or a USB (Universal Serial Bus) memory. When various drives are provided, a program can be recorded in advance on a portable recording medium, and the program can be read by the corresponding drive and executed.

[0083] In the present embodiment, a character recognition program 22, a deadline display mode list 24, a calculation formula table 26, and a starting date item list 28, which will be described later, are respectively stored in the storage unit 20. These various programs and various data can also be stored in other storage devices or external storage devices.

[0084] The deadline display mode list 24 is a list of display modes other than the date format of the deadline (refer to Figure 5 ). The display mode can be described in a regular expression. As Figure 7 shown, the calculation formula table 26 holds a pre-determined relationship between a display mode other than the date format of the deadline and a calculation formula in a table format. As Figure 8 shown, the starting date item list 28 is a list of items representing the starting date. The items representing the starting date are preset according to the type of document. Multiple items representing the starting date can also be set for one type of document.

[0085] (Step of generating a string in date format)

[0086] Next, a brief description will be given of the step of generating a string in date format.

[0087] When a keyword string representing a term is associated with a value string other than a date format, first, determine the representation pattern of the value string other than the date format. Next, determine the calculation formula corresponding to the determined representation pattern.

[0088] By pattern matching, each representation pattern in the list 24 of term representation patterns is compared with the value string other than the date format. When there is a matching representation pattern, use Figure 7 the calculation formula table 26 shown to determine the calculation formula to be used as the calculation formula corresponding to the matching representation pattern. For example, when the representation pattern is "N months from the valuation date", the following formula (1) becomes the calculation formula.

[0089] Term = $Starting Date$ + (N × 30 ± correction by month) Formula (1)

[0090] The above formula (1) adds the date of the starting date and the number of days corresponding to "N months", that is, determines the rule for calculating the term. In the above formula (1), 1 month is set to 30 days. Depending on the month, the number of days in 1 month varies within the range of 28 to 31 days, so the number of days in 1 month is corrected by month.

[0091] Next, determine the item representing the starting date. The item representing the starting date is obtained from Figure 8 the starting date item list 28 shown according to the type of the specified document. For example, if the type of the document is an estimate, the issue date, production date, and valuation date become the items representing the starting date. Although the words of the issue date, production date, and valuation date are different, they each represent the same date.

[0092] Next, determine the date of the starting date. Use the item representing the starting date as a keyword string for keyword value extraction. If a value string in date format is extracted, obtain the date represented by the extracted value string as the date of the starting date.

[0093] For example, use the issue date as a keyword string for keyword value extraction. When a value string in date format is not extracted, then use the production date as a keyword string for keyword value extraction. When a value string in date format is not extracted, then use the valuation date as a keyword string for keyword value extraction. The order of the items representing the starting date can be swapped. If a value string in date format is extracted, determine the date of the starting date.

[0094] Next, using the determined calculation formula, calculate the date of the deadline from the date of the starting date. For example, in the case where the above formula (1) is determined as the calculation formula, the starting date is determined to be the issue date and the date of the issue date is January 7, 2019, the date of the deadline of February 7, 2019 is calculated. Finally, generate a string "February 7, 2019" representing the date of the deadline in date form.

[0095] (Functional structure)

[0096] Next, the functional structure of the information processing device will be described.

[0097] Figure 9 It is a functional block diagram showing the functional structure of the information processing unit according to the first embodiment. As Figure 9 shown, the information processing unit 12 includes a character recognition unit 30, a character recognition result storage unit 32, an extraction unit 34, a form determination unit 36, a date generation unit 38, and an output unit 40.

[0098] In this embodiment, on Figure 6 the operation display unit 14 shown, Figure 2 the setting screen shown is displayed. The user operates the setting screen to specify the item representing the deadline and the type of document, thereby instructing the execution of character recognition. For example, for an estimate, specify "Estimate validity period" as the keyword string. Moreover, use Figure 6 the image reading unit 16 shown to read the image of the estimate.

[0099] The character recognition unit 30 obtains image information from the image reading unit 16 and performs character recognition on the read image. The result of the character recognition is stored in the character recognition result storage unit 32.

[0100] The extraction unit 34 obtains the specified keyword string from the operation display unit 14. The extraction unit 34 performs keyword value extraction based on the result of the character recognition on the specified keyword string and obtains the value string corresponding to the keyword string.

[0101] The form determination unit 36 determines whether the value string extracted by the extraction unit 34 is a string in date form. When a string other than the date form is extracted by the extraction unit 34, the extracted value string is transferred to the date generation unit 38. When a string in date form is extracted by the extraction unit 34, the extracted value string is transferred to the output unit 40.

[0102] The date generation unit 38 generates a string in date format representing the due date from the value string extracted by the extraction unit 34 using the result of character recognition. The date generation unit 38 transfers the generated string in date format to the output unit 40. When the date generation unit 38 cannot generate a string in date format, the date generation unit 38 transfers the value string extracted by the extraction unit 34 to the output unit 40.

[0103] The output unit 40 displays at least one of the value string extracted by the extraction unit 34 and the string generated by the date generation unit 38 as the extraction result on the operation display unit 14. For example, the output unit 40 generates screen information for displaying the extraction result and outputs it to the operation display unit 14. The operation display unit 14 displays a screen based on the screen information.

[0104] <Program>

[0105] Next, the character recognition program will be described.

[0106] Figure 10 is a flowchart showing the process flow of the character recognition program according to the first embodiment. The character recognition program is read and executed by the CPU 12A of the information processing device 10 from the storage unit 20. If the user instructs to start the character recognition process, the character recognition program is executed.

[0107] First, in step S100, the image information of the read image of the estimate is obtained from the image reading unit 16. Next, in step S102, character recognition processing is performed on the read image of the estimate. Next, in step S104, the character recognition result is stored. Next, in step S106, a string representing an item (for example, "estimate validity period") indicating the due date specified by the user is obtained as the keyword string.

[0108] Next, in step S108, using the string representing the item indicating the due date as the keyword string, keyword value extraction based on the result of character recognition is performed to extract the value string corresponding to the keyword string. Hereinafter, the string representing the item indicating the due date will be referred to as the "keyword string for the due date", and the value string corresponding to the item indicating the due date will be referred to as the "value string for the due date".

[0109] Next, in step S110, it is determined whether the value string for the due date has been extracted. When the value string for the due date has been extracted, the process proceeds to step S112. On the other hand, when the value string for the due date has not been extracted, the process proceeds to step S111. In step S111, a string indicating an error is output as the extraction result, and the routine ends.

[0110] Next, in step S112, it is determined whether the value string of the extracted deadline is in date form. When the value string is a string in a pattern of a predetermined date form, it is determined to be in date form. For example, when the value string is a date-form string in which a 4-digit number representing "year" in the Gregorian calendar, the word "year", a 1- or 2-digit number representing "month", the word "month", a 1- or 2-digit number representing "day", and the word "day" are arranged in the order of year → month → day, it is determined to be in date form. When it is a string in date form, step S132 is entered. On the other hand, when it is a string other than in date form, step S114 is entered.

[0111] Next, in step S114, it is determined whether there is a representation pattern in the deadline representation pattern list (refer to Figure 6 ) that matches the representation pattern of the value string other than in date form. When there is a matching representation pattern, step S116 is entered. On the other hand, when there is no matching representation pattern, step S134 is entered. In step S134, the value string other than in date form is output as the extraction result, and the routine ends.

[0112] Next, in step S116, using the calculation formula table (refer to Figure 6 、 Figure 7 ), the calculation formula corresponding to the matching representation pattern is determined as the calculation formula to be used.

[0113] Next, in step S118, the type of document specified by the user is obtained.

[0114] Next, in step S120, using the starting date item list (refer to Figure 6 、 Figure 8 ), the string of the item representing the starting date (e.g., "issue date") corresponding to the specified type of document is obtained as the keyword string.

[0115] Next, in step S122, using the string of the item representing the starting date as the keyword string, keyword value extraction based on the result of character recognition is performed to extract the value string corresponding to the keyword string.

[0116] Next, in step S124, it is determined whether the value string of the starting date is extracted. When the value string of the starting date is extracted, step S126 is entered. On the other hand, when the value string of the starting date is not extracted, step S134 is entered. In step S134, the value string other than in date form is extracted as the extraction result, and the routine ends.

[0117] Next, in step S126, it is determined whether the value string of the extracted starting date is in the form of a date. When it is a string in the form of a date, the process proceeds to step S128. In step S128, the date of the starting date represented by the string in the form of a date is stored. On the other hand, when it is a string other than in the form of a date, the process proceeds to step S134. In step S134, the value string other than in the form of a date is extracted as the extraction result, and the routine ends.

[0118] Next, in step S130, a date-formatted string representing the term is generated from the calculation formula determined in step S116 and the date of the starting date stored in step S128. Next, in step S132, the date-formatted string representing the term is extracted as the extraction result, and the routine ends.

[0119] In addition, when the keyword string cannot be obtained in step S120, the date represented by the string described in the pre-determined date description area within the page (for example, the upper right column of the page) can be obtained as the date of the starting date.

[0120] - Second Embodiment -

[0121] In the second embodiment, the method of generating a date-formatted string representing the term is different from that in the first embodiment. In the second embodiment, through morphological analysis of the value string of the term, the keyword string of the starting date and the string representing the period are obtained. Next, the value string corresponding to the keyword string of the starting date is obtained by keyword value extraction, and the date of the starting date is determined from the value string of the starting date. Then, the string representing the period is converted into the number of days. The date of the term is obtained from the date of the starting date and the number of days.

[0122] <Information Processing Apparatus>

[0123] Figure 11 is a block diagram showing the structure of the storage unit of the information processing apparatus according to the second embodiment. As Figure 11 shown, in this embodiment, a character recognition program 22, a starting date item list 28, a morphological analysis dictionary 50, and a number-of-days conversion rule 52, which will be described later, are respectively stored in the storage unit 20. These various programs and various data can also be stored in other storage devices or external storage devices.

[0124] The morphological analysis dictionary 50 is dictionary data for the above-mentioned morphological analysis. The number-of-days conversion rule 52 is a rule for converting the string B representing the period into the number of days. As Figure 16 shown, in the number-of-days conversion rule 52, for example, "month" is 30 days, "week" is 7 days, "day" is 1 day, etc., and the number of days represented by the classifier is determined in advance. For example, when it is "3 weeks", the number of days of 21 days is calculated by multiplying 7 days by 3 times from the numeral "3" and the classifier "week".

[0125] <Procedure>

[0126] Next, the character recognition program will be described.

[0127] Figure 12 It is a flowchart showing the processing flow of the character recognition program according to the second embodiment. The character recognition program is read out and executed by the CPU 12A of the information processing apparatus 10 from the storage unit 20. When the user instructs to start the character recognition process, the character recognition program is executed.

[0128] First, in step S200, the image information of the read image of the estimate is acquired from the image reading unit 16. Next, in step S202, character recognition processing is performed on the read image of the estimate. Next, in step S204, the character recognition result is stored. Next, in step S206, a character string representing an item of the deadline specified by the user is acquired as a keyword string.

[0129] Next, in step S208, using the character string representing the item of the deadline as the keyword string, keyword value extraction based on the result of character recognition is performed, and the value string corresponding to the keyword string is extracted. Next, in step S210, it is determined whether the value string of the deadline has been extracted. When the value string of the deadline has been extracted, the process proceeds to step S212. On the other hand, when the value string of the deadline has not been extracted, the process proceeds to step S211. In step S211, a character string representing an error is output as the extraction result, and the routine ends.

[0130] Next, in step S212, it is determined whether the extracted value string of the deadline is in date format. When it is a date format string, the process proceeds to step S228. On the other hand, when it is a string other than the date format, the process proceeds to step S214.

[0131] Next, in step S214, morphological analysis of the extracted value string of the deadline is performed. As described above, using the morphological analysis dictionary 50, the extracted value string of the deadline is divided into units of morphemes, and each morpheme is discriminated, assigned a part of speech, conjugation, reading, etc.

[0132] Next, in step S216, it is determined whether the form of the value string of the deadline conforms to the first form. When it conforms to the first form, the process proceeds to step S218. On the other hand, when it does not conform to the first form, the process proceeds to step S232.

[0133] Next, in step S218, the analysis process of string A is executed.

[0134] (Analysis process of string A)

[0135] Here, the "analysis and processing of string A" will be described.

[0136] Figure 13 is a flowchart showing the process of the analysis and processing of string A. In the analysis and processing of string A, first, in step S300, a string representing the starting date is extracted from string A.

[0137] Next, in step S302, the string of the item representing the starting date is used as a keyword string, and keyword value extraction based on the character recognition result is performed to extract the value string corresponding to the keyword string.

[0138] Next, in step S304, it is determined whether a value string of the starting date has been extracted. When a value string of the starting date is extracted, go to step S306. On the other hand, when a value string of the starting date has not been obtained, go to step S310. In step S310, the Figure 14 shown "date extraction process of the starting date" is executed, and the routine ends.

[0139] Next, in step S306, it is determined whether the extracted value string of the starting date is in date format. When it is a string in date format, go to step S308. In step S308, the date of the starting date represented by the string in date format is stored, and the routine ends. On the other hand, when it is a string other than the date format, the routine ends.

[0140] In addition, when a keyword string cannot be obtained in step S300, the date represented by the string described in the predetermined date description area within the page can also be obtained as the date of the starting date.

[0141] (Date extraction process of the starting date)

[0142] Here, the "date extraction process of the starting date" will be described.

[0143] Figure 14 is a flowchart showing the process of the date extraction process of the starting date. In the date extraction process of the starting date, first, in step S400, the type of document specified by the user is obtained.

[0144] Next, in step S402, using the starting date item list (refer to Figure 6 、 Figure 8 ), a string of the item representing the starting date (for example, "issue date") corresponding to the specified type of document is obtained as a keyword string.

[0145] Next, in step S404, the string of the item representing the starting date is used as a keyword string, and keyword value extraction based on the character recognition result is performed to extract the value string corresponding to the keyword string.

[0146] Next, in step S406, it is determined whether a value string of the starting date is extracted. When the value string of the starting date is extracted, step S408 is entered. On the other hand, when the value string of the starting date is not extracted, the routine ends.

[0147] Next, in step S408, it is determined whether the extracted value string of the starting date is in date format. When it is a string in date format, step S410 is entered. In step S410, the date of the starting date represented by the string in date format is stored. On the other hand, when it is a string other than in date format, the routine ends.

[0148] In addition, when the keyword string cannot be obtained in step S402, the date represented by the string described in the predetermined date description area within the page can be obtained as the date of the starting date.

[0149] Return to Figure 12 the description of. Next, in step S220, it is determined whether the result of the analysis process of string A has obtained the date of the starting date. When the date of the starting date is obtained, step S222 is entered. When the date of the starting date is not obtained, step S230 is entered. In step S230, a value string other than in date format is output as the extraction result, and the routine ends.

[0150] Next, in step S222, the analysis process of string B is executed.

[0151] (Analysis process of string B)

[0152] Here, the "analysis process of string B" is described.

[0153] Figure 15 is a flowchart showing the process of the analysis process of string B. In the analysis process of string B, first, in step S500, the string B representing the period is converted into days using the number-of-days conversion rule. Next, in step S502, the calculated number of days is stored, and the routine ends.

[0154] Return to Figure 12 the description of. Next, in step S224, it is determined whether the result of the analysis process of string B has obtained the number of days. When the number of days is obtained, step S226 is entered. When the number of days cannot be obtained, step S230 is entered. In step S230, a value string other than in date format is output as the extraction result, and the routine ends.

[0155] Next, in step S226, a string in the date format representing the deadline is generated from the date of the starting date obtained in step S218 and the number of days obtained in step S222. Next, in step S228, the string in the date format representing the deadline is output as the extraction result, and the routine ends.

[0156] Moreover, when in step S216, the form of the value string of the deadline does not conform to the first form and the process proceeds to step S232, then in step S232, it is determined whether the form of the value string of the deadline conforms to the second form. When it conforms to the second form, the process proceeds to step S234. On the other hand, when it does not conform to the second form, the process proceeds to step S234.

[0157] Next, in step S234, the Figure 15 analysis process of the string B shown is executed.

[0158] Return to Figure 12 the description. Next, in step S236, it is determined whether the number of days is obtained as a result of the analysis process of the string B. When the number of days is obtained, the process proceeds to step S238. When the number of days cannot be obtained, the process proceeds to step S230. In step S230, a value string other than the date format is output as the extraction result, and the routine ends.

[0159] In step S238, the Figure 14 "starting date extraction process" shown is executed.

[0160] Next, in step S240, it is determined whether the starting date is obtained as a result of the starting date extraction process. When the starting date is obtained, the process proceeds to step S242. When the starting date cannot be obtained, the process proceeds to step S230. In step S230, a value string other than the date format is output as the extraction result, and the routine ends.

[0161] Next, in step S242, a string in the date format representing the deadline is generated from the number of days obtained in step S234 and the starting date obtained in step S238. Next, in step S244, the string in the date format representing the deadline is output as the extraction result, and the routine ends.

[0162] -Variant Example-

[0163] In addition, the structures of the information processing apparatus and the program described in the above embodiments are examples, and it is natural that their structures can be changed without departing from the spirit of the present invention.

[0164] In each of the above-described embodiments, an example in which each of the processes is implemented by a software structure by executing a program using a computer has been described, but the present invention is not limited thereto. For example, each of the processes may also be implemented by a hardware structure or a combination of a hardware structure and a software structure.

[0165] The above-described embodiments of the present invention are provided for purposes of illustration and description. In addition, the embodiments of the present invention do not comprehensively and exhaustively include the present invention, and do not limit the present invention to the disclosed manner. Obviously, various modifications and variations are apparent to those skilled in the art to which the present invention pertains. The present embodiment is selected and described in order to most easily illustrate the principles of the present invention and its applications. Thus, other technicians in the art can understand the present invention through various modified examples that are optimized for specific uses assumed to be various embodiments. The scope of the present invention is defined by the above claims and their equivalents.

Claims

1. An information processing apparatus, comprising: a first extraction unit that extracts, from a result of character recognition of a read image of a document including items and values corresponding to the items, a value corresponding to a first item that is an item representing a deadline; a generation unit that, when the value extracted by the first extraction unit is a first string other than a date format, generates a second string in a date format representing the date of the deadline based on the character recognition result; and an output unit that outputs the generated second string as a value corresponding to the first item.

2. The information processing apparatus according to claim 1, wherein the generation unit includes: an acquisition unit that acquires a second item different from the first item; and a second extraction unit that extracts, from the character recognition result, a value corresponding to the second item, and generates the second string in the date format based on the value extracted corresponding to the acquired second item.

3. The information processing apparatus according to claim 2, wherein the second item is an item representing a starting date that is a reference for calculating the deadline, and when the value extracted by the second extraction unit is a third string in a date format, the generation unit acquires the date represented by the third string as the date of the starting date.

4. The information processing apparatus according to claim 3, wherein when the value extracted by the second extraction unit is a fourth string other than a date format, the generation unit acquires the date represented by a fifth string described in a predetermined date description area of the image as the date of the starting date.

5. The information processing apparatus according to claim 3 or 4, wherein the acquisition unit acquires, from an item storage unit, an item representing the starting date corresponding to the type of the specified document, and the item storage unit stores in advance items representing the starting date according to the types of a plurality of documents.

6. The information processing apparatus according to claim 3 or 4, wherein the acquisition unit acquires the item representing the starting date by morpheme analysis of the first string.

7. The information processing apparatus according to claim 3 or 4, wherein the generation unit uses a calculation formula for calculating the date of the deadline from the date of the starting date, acquires the date of the deadline corresponding to the acquired date of the starting date, and generates the second string in the date format.

8. The information processing apparatus according to claim 7, wherein the calculation formula for calculating the date of the deadline is acquired from a calculation formula storage unit according to the representation pattern of the first string, and the calculation formula storage unit stores in advance calculation formulas according to a plurality of representation patterns other than the date format.

9. The information processing apparatus according to claim 3 or 4, wherein the generation unit acquires, by morpheme analysis of the first string, the period from the starting date to the deadline, converts the acquired period into days, and acquires the date of the deadline corresponding to the number of days and the date of the starting date, thereby generating the second string in the date format.

10. The information processing apparatus according to any one of claims 1 to 4, wherein The output unit displays the second string in the date format on the display unit.

11. The information processing apparatus according to claim 10, wherein the output unit replaces the first string other than the date format included in the result of the character recognition with the second string in the date format.

12. A recording medium that records a program for causing a computer to function as each unit of the information processing apparatus according to any one of claims 1 to 11.

13. An information processing method, comprising the steps of: a first extraction step of extracting, from the result of character recognition of a read image of a document including items and values corresponding to the items, a value corresponding to a first item that is an item representing a deadline; a generation step of generating, when the value extracted by the first extraction step is a first string other than the date format, a second string in the date format representing the date of the deadline according to the result of the character recognition; and an output step of outputting the generated second string as the value corresponding to the first item.

Images